4 ms·

So what is closer to Bob: that encrypted file or an empty file? On the one hand, the encrypted file is only a couple of keystrokes away from creating Bob. But t

So what is closer to Bob: that encrypted file or an empty file? On the one hand, the encrypted file is only a couple of keystrokes away from creating Bob. But there is no known process to find the right keystrokes. The empty file needs a lot more keystrokes to create Bob (rewriting it from scratch), but there is a known process for doing it. Edit: Kids, I'll just assume whoever downvotes me doesn't like Computer Science very much. That's OK, but what is with the intolerance?

mahouse 11y ago
User name relevant.
- facepalm 11y ago
  What do you mean?
DanBC 11y ago
(I didn't downvote you) Assuming the OP is telling the truth and the password was just someone mashing the keys: it'd make a nice project for someone to try to see if it's vulnerable to cryptanalysis. Windows XP was released around 2001. MS have had a few cryptographic errors.
- facepalm 11y ago
  That adds another twist :-) I just thought it would be amusing to think about different metrics and human perception. In the metric of keystrokes needed the encrypted file is closer to Bob, whereas in the metric of time needed the empty file might be closer. Human perception intuitively thinks the encrypted file is closer, but it might be wrong.
ikeboy 11y ago
The entropy of Bob given the encrypted file is less than the entropy of Bob given an empty file. So Bob is closer to the encrypted file.
- facepalm 11y ago
  You mean if you have the file you need less information to retrieve Bob than if you don't have the file? That's certainly a valid way of looking at it.
mikeash 11y ago
I vote for the encrypted file being closer. Yes, there is a known process for going from an empty file to something like Bob, but you'll never reproduce it 100% exactly. Plus, the encryption they used probably sucks.
dtech 11y ago
It depends entirely on how you define "closer" (so you have to define distance). It's hamming distance should be about equivalent, if you assume Microsoft Bob contains about 50% bits set to one. Altough I have no idea if that assumption is valid for a typical Win32 binary.

mahouse 11y ago

User name relevant.

facepalm 11y ago
What do you mean?

DanBC 11y ago

(I didn't downvote you) Assuming the OP is telling the truth and the password was just someone mashing the keys: it'd make a nice project for someone to try to see if it's vulnerable to cryptanalysis. Windows XP was released around 2001. MS have had a few cryptographic errors.

facepalm 11y ago
That adds another twist :-) I just thought it would be amusing to think about different metrics and human perception. In the metric of keystrokes needed the encrypted file is closer to Bob, whereas in the metric of time needed the empty file might be closer. Human perception intuitively thinks the encrypted file is closer, but it might be wrong.

ikeboy 11y ago

The entropy of Bob given the encrypted file is less than the entropy of Bob given an empty file. So Bob is closer to the encrypted file.

facepalm 11y ago
You mean if you have the file you need less information to retrieve Bob than if you don't have the file? That's certainly a valid way of looking at it.

mikeash 11y ago

I vote for the encrypted file being closer. Yes, there is a known process for going from an empty file to something like Bob, but you'll never reproduce it 100% exactly. Plus, the encryption they used probably sucks.

dtech 11y ago

It depends entirely on how you define "closer" (so you have to define distance). It's hamming distance should be about equivalent, if you assume Microsoft Bob contains about 50% bits set to one. Altough I have no idea if that assumption is valid for a typical Win32 binary.